full-review

Dispatch independent subagents to review code and merge severity-rated findings.

Updated May 4, 2026

Search Tags:#code review #parallel processing #subagents #severity rating #comprehensive evaluation

Ready to run

npx skills add https://github.com/cjohnhanson/almanac --skill full-review-cjohnhanson

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: full-review
Source: https://github.com/cjohnhanson/almanac/tree/main/skills/full-review
Command: npx skills add https://github.com/cjohnhanson/almanac --skill full-review-cjohnhanson

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Manages a parallel code review process, leveraging multiple review skills to provide a thorough and detailed evaluation of code changes.

Core Features & Use Cases

Parallel Subagent Reviews: Each review skill operates as an independent subagent, examining code changes without context sharing.
Severity-Rated Findings: Reports on issues with severity ratings, including file:line locations and suggested fixes.
End-to-End Review: Merges findings from all subagents into a single severity-ranked list for the user.

Quick Start

Run /full-review for a comprehensive review of your code changes.

Frequently Asked Questions about full-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I run a comprehensive code review on a large codebase?▼

Run a comprehensive code review by dispatching multiple independent subagents in parallel, each performing a specific review task without shared context. This approach evaluates large codebases thoroughly by merging all findings into a single severity-ranked list with file:line locations and suggested fixes.

What is the best way to get severity ratings for code review findings?▼

Severity ratings for code review findings are generated by independent subagents that evaluate code changes and report issues. Each subagent assigns a severity rating, includes file:line locations, and provides suggested fixes, which are then merged into a single severity-ranked list.

Can I use parallel subagents to review code changes without shared context?▼

Yes, parallel subagents can review code changes without shared context. Each subagent operates independently on the same codebase, performing its specific review task and reporting findings without relying on the context of other subagents, ensuring an unbiased comprehensive evaluation.

Do I need multiple review skills to perform an end-to-end code review?▼

Yes, you need multiple review skills to perform an end-to-end code review using this approach. The system leverages these skills as independent subagents to provide a thorough and detailed evaluation of code changes across large codebases.

How does parallel processing improve code evaluation for large codebases?▼

Parallel processing improves code evaluation for large codebases by dispatching multiple independent subagents simultaneously. Each subagent handles a specific review task, reducing overall review time while providing a thorough evaluation merged into a severity-ranked list of findings.